Dissertations / Theses on the topic '1H-NMR metabolomics'

El perfilat de sèrum per ressonància magnètica nuclear de protó (1H-RMN) està especialment indicat per a anàlisi a gran escala en estudis epidemiològics, nutricionals o farmacològics. L’espectroscòpia 1H-RMN requereix mínima manipulació de mostra i gràcies a la seva resposta quantitativa permet la comparació directa entre laboratoris. Un perfilat complet de sèrum per 1H-RMN requereix de tres mesures que es corresponen amb tres espècies moleculars diferents: lipoproteïnes, metabòlits de baix pes molecular i lípids. El perfilat de sèrum per 1H-RMN permet obtenir informació de grandària, nombre de partícules i contingut lipídic de les subfraccions lipoproteiques, així com l'abundància d'aminoàcids, productes de la glicòlisi, cossos cetònics, àcids grassos i fosfolípids, entre d'altres. No obstant això, la complexitat espectral afavoreix la inclusió d'errors en l'anàlisi manual de les dades, mentre que les múltiples interaccions moleculars en el sèrum comprometen la seva precisió quantitativa. És per tant necessari desenvolupar mètodes robustos de perfilat metabòlic per consolidar la 1H-RMN en la pràctica clínica. Per a això, aquesta tesi presenta diverses estratègies metodològiques i computacionals. En el primer treball, es van desenvolupar mètodes de regressió dels lípids del perfil lipídic clàssic, generalitzables a mostres de població sana i amb valors de lípids i lipoproteïnes anormals. Aquests lípids representen els principals indicadors de risc cardiovascular i els objectius terapèutics primaris. En el segon estudi caracteritzem els errors de quantificació en el perfilat 1H-RMN de metabòlits clínicament rellevants, que són deguts a la seva agregació a la proteïna sanguínia. També proposem un mètode que fomenta la competició per l'agregació i que ens permet obtenir quantificacions dels nostres metabòlits properes a les absolutes. Finalment, el tercer treball presenta LipSpin: una eina bioinformàtica de codi obert específicament dissenyada per al perfilat de lípids per 1H-RMN. A més, aquest estudi exposa alguns aspectes metodològics per millorar l'anàlisi de lípids per RMN.
El perfilado de suero por resonancia magnética nuclear de protón (1H-RMN) está especialmente indicado para el análisis a gran escala en estudios epidemiológicos, nutricionales o farmacológicos. La espectroscopía 1H-RMN requiere mínima manipulación de muestra y gracias a su respuesta cuantitativa permite la comparación directa entre laboratorios. Un perfilado completo de suero por 1H-RMN requiere de tres mediciones que se corresponden con tres especies moleculares distintas: lipoproteínas, metabolitos de bajo peso molecular y lípidos. El perfilado de suero por 1H-RMN permite obtener información de tamaño, número de partículas y contenido lipídico de las subfracciones lipoproteicas, así como la abundancia de aminoácidos, productos de la glicólisis, cuerpos cetónicos, ácidos grasos y fosfolípidos, entre otros. Sin embargo, la complejidad espectral favorece la inclusión de errores en el análisis manual de los datos, mientras que las múltiples interacciones moleculares en el suero comprometen su precisión cuantitativa. Es por tanto necesario desarrollar métodos robustos de perfilado metabólico para consolidar la 1H-RMN en la práctica clínica. Para ello, esta tesis presenta varias estrategias metodológicas y computacionales. En el primer trabajo, se desarrollaron métodos de regresión de los lípidos del perfil lipídico clásico, generalizables a muestras de población sana y con valores de lípidos y lipoproteínas anormales. Estos lípidos representan los principales indicadores de riesgo cardiovascular y los objetivos terapéuticos primarios. En el segundo estudio caracterizamos los errores de cuantificación en el perfilado 1H-RMN de metabolitos clínicamente relevantes, que son debidos a su agregación a la proteína sanguínea. También proponemos un método que fomenta la competición por la agregación y que nos permite obtener cuantificaciones de nuestros metabolitos cercanas a las absolutas. Por último, el tercer trabajo presenta LipSpin: una herramienta bioinformática de código abierto específicamente diseñada para el perfilado de lípidos por 1H-RMN. Además, este estudio expone algunos aspectos metodológicos para mejorar el análisis de lípidos por RMN.
1H-NMR serum profiling is especially suitable for high-throughput epidemiological studies and large-scale nutritional studies and drug monitoring. It requires minimal sample manipulation and its quantitative response allows inter-laboratory comparison. A comprehensive 1H-NMR serum profiling consists of three measurements encoding different molecular species: lipoproteins, low-molecular-weight metabolites and lipids. 1H-NMR serum profiling provides information of size, particle number and lipid content of lipoprotein subclasses, as well as abundance of amino acids, glycolysis-related metabolites, ketone bodies, fatty acids and phospholipids, among others. However, the spectral complexity promotes errors in manual data analysis and the multiple molecular interactions within the sample compromise reliable quantifications. Developing robust methods of metabolite serum profiling is therefore desirable to consolidate high-throughput 1H-NMR in the clinical practice. This thesis presents several methodological and computational strategies to that end. In the first study, we developed generalizable regression methods for lipids in routine clinical practice (known as “lipid panel”), to be applied in healthy population and in a wide spectrum of lipid and lipoprotein abnormalities. These standard lipids are still the main measurements of cardiovascular risk and therapy targets. In the second study, we characterised the quantitative errors introduced by protein binding in 1H-NMR profiling of clinically-relevant LMWM in native serum. Then, we proposed a competitive binding strategy to achieve quantifications closer to absolute concentrations, being fully compatible with high-throughput NMR. Finally, the third study presents LipSpin: an open source bioinformatics tool specifically designed for 1H-NMR profiling of serum lipids. Moreover, some methodological aspects to improve NMR-based serum lipid analysis are discussed.

Orientador: Anita Jocelyne Marsaioli
Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Química
Made available in DSpace on 2018-08-21T04:07:12Z (GMT). No. of bitstreams: 1 Lopes_ThiagoInacioBarros_M.pdf: 3472847 bytes, checksum: 6ec44b5e5d2ad869887d1d1a884a0682 (MD5) Previous issue date: 2012
Resumo: A organização mundial de saúde (OMS) estima que existam atualmente mais de 1,5 bilhões de adultos com sobrepeso, número que é esperado dobrar até 2015. Obesidade e sobrepeso são enfermidades caracterizadas pelo acúmulo excessivo de gordura corporal e têm sido associadas a vários problemas de saúde. Vários fármacos têm sido utilizados no tratamento desta enfermidade, entre os mais utilizados se encontra o Orlistate, um inibidor de lipases gástricas utilizado na redução da absorção de gordura da dieta, auxiliando na perda e manutenção do peso. Esta dissertação teve como objetivo principal avaliar as alterações metabólicas sofridas por pacientes com sobrepeso tratados com Orlistate por 120 dias. Para tornar mais didático, o trabalho foi divido em duas Partes: Parte I, implementação da metodologia analítica para análise de ácidos graxos por CG-EM e avaliação do perfil de ácidos graxos em indivíduos com sobrepeso tratados com Orlistate; e Parte II, avaliação do perfil metabólico de plasma por RMN de ¹H de indivíduos com sobrepeso tratados com Orlistate. Na primeira etapa do trabalho, análise de componentes principais foi aplicada na seleção íons (m/z) para quantificação de ácidos graxos, após preparação de ésteres metílicos correspondentes, por CG-EM usando monitoramento de íons selecionados. Quatro íons foram então selecionados de forma a aumentar a detectividade sem perda completa de informação qualitativa. Os íons de m/z 74, 79, 81 e 87 foram selecionados e permitiram a quantificação de vários ácidos graxos, além da determinação do número de insaturações dos mesmos relacionando a abundância relativa dos íons à presença de insaturações. A metodologia analítica implementada permitiu quantificar ácidos graxos esterificados em vários lipídios presentes no sangue, após transesterificação para produção dos ésteres metílicos de ácidos graxos, com adequada precisão, repetitividade e baixos limites de detecção e quantificação. Na segunda etapa, a metodologia analítica foi aplicada no estudo do perfil de ácidos graxos de 20 mulheres com sobrepeso tratadas com Orlistate durante 120 dias. O tratamento não reduziu o índice de massa corpórea contribuiu para diminuição significativa dos níveis de colesterol-HDL no plasma e do conteúdo de colesterol na membrana de eritrócitos, além de alterar as proporções relativas de vários ácidos graxos essenciais e exógenos em vários lipídios estudados. Adicionalmente foi observado um perfil de ácido graxo significativamente diferente para os indivíduos magros (controles) em comparação com indivíduos com sobrepeso. Na última etapa, o perfil metabólico de plasma por RMN de ¹H foi estudado por uma abordagem metabolômica. A análise discriminatória por quadrados mínimos parciais (PLS-DA) revelou que alterações nos níveis de lactato e magnésio são importantes na diferenciação entre indivíduos com sobrepeso tratados e não-tratados com Orlistate, sinalizando diminuição destes metabólitos relacionada ao tratamento. Não há relatos anteriores da alteração dos níveis de lactato devido ao tratamento. Adicionalmente, os níveis de triacilglicerídios, alanina e lactato contribuíram significativamente na distinção entre indivíduos magros e com sobrepeso
Abstract: The World Health Organization (WHO) estimates that there are currently more than 1.5 billion overweight adults, a number that is expected to double until 2015. Obesity and overweight are diseases characterized by excessive accumulation of body fat and linked to various health problems. Several drugs have been used to treat such diseases. Orlistat is a gastric lipase inhibitor largely used to reduce the absorption of dietary fat, helping weight loss. Thus this thesis aimed at the metabolic variation of overweight subjects treated with Orlistat for 120 days. This work is divided into two parts: Part I, implementation of the analytical methodology for analysis of fatty acids by GC-MS and evaluation of the fatty acid profile in overweight subjects treated with Orlistat; and Part II, evaluation of the metabolic profile of plasma of overweight subjects treated with Orlistat using ¹H NMR. In the Part I of this work, principal component analysis was applied to selected ions (m/z) for determination of fatty acids, after preparation of the corresponding methyl esters, by GC-MS using selected ion monitoring. Four ions were selected. The ions of m/z 74, 79, 81 and 87 allowed the quantification of various fatty acids and determination of the number of double bounds in theo fatty acids through the relative abundance of these selected ions. The analytical methodology implemented permitted quantify esterifing fatty acids in various lipids in blood, after transesterification for production of methyl esters of fatty acids, with adequate accuracy, repeatability and low limits of detection and quantification. In the second step, the analytical methodology was applied to study the fatty acid profile of 20 overweight women treated with Orlistat for 120 days. Although there was no significant reduction in body mass index, the treatment contributed to significant reduction of HDL-cholesterol levels in plasma and the cholesterol content in erythrocyte membranes. The Orlistat also alters the relative proportions of various fatty acids in the several lipids studied. Additionally, a significantly different fatty acid profile was observed for lean subjects (controls) compared to overweight subjects. In Part II, the metabolic profile of plasma obtained by ¹H NMR was studied by a metabolomics approach. The partial least squares discriminant analysis (PLS-DA) revealed changes in lactate and magnesium level, these metabolites were important to differentiating between overweight subjects treated and not-treated with Orlistat, suggesting that the level of these metabolites decreased with treatment. There are no previous reports of changes in lactate levels. Additionally, levels of triglycerides, alanine and lactate were highlighted by PLS-DA into distinguishing between lean and overweight individuals
Mestrado
Quimica Organica
Mestre em Química

This study has characterized beer samples taken from two different Finnish beer: Claudia’s Citra Stout (Mustialan Kartano Panimo) and Thunder Chief (Hopping Brewsters Panimo). Both of them have been analyzed from several steps of the brewing process, in order to characterize the final product (4.) and the metabolites change (1. after mashing, 2. after Whirlpool and 3. after secondary fermentation). The aim of this research was to discover how each compound changed during the beer making process, how this could be useful to characterize different styles of beers and if there was any by-product, in the liquids extract discarded, that could be valorized. The beermaking process has been followed from a metabolomics perspective, based on 1H NMR (Nuclear Magnetic Resonance) . In order to get a representative result, 33 NMR spectra have been analyzed, which we obtained by experimenting with different samples preparation methods [(a),(b),(c),(d)], to get the best characterization possible. The majority of the metabolites identified decreased in concentration from the steps before fermentation to post-secondary fermentation. In particular, it has been possible to distinguish 73 metabolites in unfiltered and unpasteurized beer (artigianal), and in commercial beer (Viktor), which has been used as reference. All of them were different in quantitative terms and through their characterization it was possible to identify exactly their presence in others beer samples. Furthermore was possible to obtain an objective opinion on the specific molecules and metabolisms acting on them, in order to optimize a recipe or to modify one that presents problems. Just for this last purpose, it was useful to verify changes that occur during the main steps (1-2-3-4), in order to identify not only the problem, but also in which step this is presented, and consequently how to intervene.

Orientador: Anita Jocelyne Marsaioli
Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Quimica
Made available in DSpace on 2018-08-13T13:31:18Z (GMT). No. of bitstreams: 1 Martins_LucasGelain_M.pdf: 1200827 bytes, checksum: 8708dd4e868d8749ef0f4a5d27b03248 (MD5) Previous issue date: 2009
Resumo: Técnicas de RMN de H foram aplicadas no estudo de metabolômica de Theobroma cacao e interações proteínas ¿ ligantes. Esses dois tópicos são reportados em dois capítulos. O primeiro descreve a análise quantitativa por RMN de H de 8 metabolitos de Theobroma cacao endógenos ou produzidos durante a fermentação de 10 diferentes variedades resistentes ao fungo Moniliophthora perniciosa. Os espectros foram obtidos de extratos aquosos saturando o sinal da água com pressaturação. Os metabólitos monitorados foram: glicose, sacarose, frutose, etanol, cafeínas e os ácidos acético, lático e succínico. O teor dos metabólitos presentes em 7 dessas variedades são similares àqueles de T. cacao comum e análise sensorial selecionou aquelas com maior teor de cafeína como as melhores. O segundo capítulo descreve as interações acetilcolinesterase/alcalóides de Amaryllidaceae e proteína de membrana bacteriana/fosfomicina, com os mapas de epitopo de interação dos ligantes com as macromoléculas determinados por RMN de H ¿ STD (Saturation Transfer Difference). O mapa de epitopo para acetilcolinesterase/fisostigmina foi confirmado por cálculos de ¿doking¿ molecular. A constante de dissociação aparente da fisostigmina com acetilcolinesterase foi obtida através das medidas de T1 seletivo. Interação entre células integras de Escherichia coli CCT 5050, Serratia liquefaciens CCT 7262 e fosfomicina indicaram que microrganismo resistente ao antibiótico (Serratia liquefaciens) não apresentaram sinais no experimento de RMN de H ¿ STD enquanto que o microrganismo não resistente (Escherichia coli) apresenta sinais no mesmo experimento. Esses resultados certamente contribuirão para a elucidação do mecanismo de ação da fosfomicina
Abstract: H NMR tecniques were applied to study Theobroma cacao metabolomics and protein-ligand supramolecular interactions. These two topics are reported in two chapters. First chapter describes the H NMR quantitative analysis of 8 Theobroma cacao secondary metabolites which are endogenous or produced during the fermentation process of 10 different varieties resistent to the Moniliophthora perniciosa fungus. The spectra were obtained from the aqueous extracts saturating the water signal with the PRESAT tecnique. The monitored metabolites were: glucose, sacharose, fructose, ethanol, cafeine and the acids acetic, latic and succinic. The abundance of the metabolites present in 7 of these varieties were similar to those in comon T. Cacao and sensory analysis selected those with high cafeine content as the best. The second chapter describes the supramolecular interactions of acetylcholisterase/Amaryllidaceae alkaloids and bacterial membrane protein /fosfomicine, with the epitope mapping of the interactions of the ligands to the macromolecules obtained by H NMR STD (saturation transfer difference). The epitope mapping of acetylcholisterase/Amaryllidaceae alkaloids was confirmed by molecular docking calculations. The apparent dissociation constant of the fisostigmine which is an acetylcholinesterase inhibitor was obtained by applying selective T1. Interactions of Escherichia coli CCT 5050, Serratia liquefaciens CCT 7262 whole cells and fosfomycin indicated that microorganisms resistent to this antibiotic (Serratia liquefaciens) did not show signal in the H NMR STD experiment while non resistent microorganism (Escherichia coli) showed signals in the same experiment. This data will certainly contribute to the elucidation of fosfomycin action mechanism
Mestrado
Quimica Organica
Mestre em Química

In questo lavoro di tesi sono stati analizzati campioni di orate (Sparus aurata) da acquacoltura allevate con differenti diete, commerciali o innovative, con lo scopo di verificare se le diverse formulazioni dei mangimi hanno un effetto sul profilo molecolare dei pesci. A tal proposito, è stata utilizzata la spettroscopia di Risonanza Magnetica Nucleare (1H NMR) accoppiata ad un approccio metabolomico poiché, rispetto ad altre tecniche, consente la determinazione simultanea di un’ampia classe di metaboliti che caratterizzano il campione attraverso la generazione di spettri che contengono un ampio spettro di informazioni. Questo lavoro di tesi si sposa bene con l’attuale tendenza di sviluppo dell’acquacoltura, che viene vista come uno strumento interessante per fornire cibo e migliorare la sicurezza alimentare, ma allo stesso tempo esso è in accordo anche con la sempre più elevata attenzione verso sostenibilità, qualità e conformità alle richieste dei consumatori, che sono oggi le principali tendenze nel settore alimentare. Negli ultimi anni si è assistito ad un ampio sviluppo dell’acquacoltura e questo ha portato alla necessità di ricercare strategie per migliorare l’efficienza, la produttività, ma anche la sostenibilità di questi sistemi. Questo ha condotto ad un elevato aumento delle ricerche riguardanti questi aspetti dell’acquacoltura, e l’impatto che la dieta ha sul metabolismo dei pesci è uno di questi. I dati spettroscopici sono stati interpretati attraverso analisi statistica multivariata, nello specifico attraverso l’analisi delle componenti principali, PCA, per individuare eventuali differenze significative tra i tre gruppi a seguito della diversa tipologia di alimentazione. Dal presente lavoro di tesi è stato possibile concludere che i campioni vengono distribuiti e classificati in funzione della dieta. Inoltre, è stato possibile individuare i metaboliti che maggiormente determinano questa separazione.

NMR-linked metabolomics analysis was employed to investigate urinary and human plasma profiles collected from Niemann Pick type C1 disease patients (NP-C1), in addition to aqueous extracts of liver samples of an NP-C1 mouse model. NP-C1 is a lysosomal storage disorder caused by mutations in the lysosomal proteins NPC1 and NPC2, which are involved in lysosomal cholesterol trafficking. NP-C1 disease is a fatal genetic disorder, characterised by neurodegeneration and hepatic damage. Miglustat (MGS) is the only approved drug for this disease, and consequently, plasma and urine samples collected from MGS-treated patients were also investigated. The ability of 1H NMR analysis to detect a wide range of metabolites simultaneously served to characterize the metabolic profiles of urine, plasma and hepatic tissue samples investigated in order to perform linked multivariate analysis (MVA). Additionally, MGS was identified in urine samples collected from NP-C1 treated patients. MVA employing both parametric and machine learning-based techniques was conducted to classify samples according to their disease status, and also to seek biomarkers that could aid in the diagnosis and/or prognosis of the disease. Moreover, a new technique was introduced in a metabolomics context, Correlated Component Regression (CCR), and the suitability of Random Forests (RFs) for variable selection was also explored. We were able to differentiate urine samples collected from NP-C1 patients from those collected from heterozygous controls, and also propose several metabolites as NP-C1 urinary biomarkers such as bile acids, 2-hydroxy-3-methylbutyrate, 3-aminoisobutyrate, 5-aminovalerate, trimethylamine, methanol, creatine and quinolinate. The 1H NMR linked metabolomics study of plasma samples revealed major distinctions among the groups investigated, metabolic alterations ascribable to the disease pathology were mainly observed as changes in the lipoprotein profiles of NP-C1 patients. Hepatic tissue extracts analysed revealed major disturbances in amino acid metabolism, along with impairments in the NAD+/NADH production and redox status. Gut microbiota and bile acid metabolism were also highlighted as features altered in NP-C1 disease. CCR linked to Linear Discriminant Analysis was evaluated as a new tool for metabolomics analysis, giving accurate results when compared to alternative techniques tested. Additionally, the suitability of Random Forests and associated recursive feature elimination for variable selection in metabolomics studies was contrasted, suggesting that those strategies relying on a variable ranking to select the top features for discrimination are more suitable for metabolomics investigations than those that iteratively remove a percentage of the least effective features until the classification performance decays.

RMN és una plataforma analítica utilitzada per quantificar els metabòlits presents en les mostres de metabolòmica. Els espectres de 1H-RMN mostren múltiples senyals de metabòlits amb tres paràmetres específics (desplaçament químic, ample mitjà de banda, intensitat) que poden mostrar reactivitat a les condicions de la mostra. Aquesta reactivitat perjudica l'optimització del fitat dels espectres necessari per a realitzar el perfilat automàtic de metabòlits de les mostres. L'objectiu d'aquesta tesi va ser l'exploració de l'ús de tècniques de tendència basades en Machine Learning (ML) amb l'ús de fluxos de treball robustos per modelar i explotar la informació present en els diferents paràmetres de senyal durant el perfilat de metabòlits dels conjunts de dades 1H-NMR. En particular, les aplicacions considerades van ser la millora de la classificació de les mostres en els estudis de metabolòmica i la millora de la qualitat del perfilat automàtic. A més d'assolir aquests objectius, també es van obtenir èxits addicionals (per exemple, la generació d'una nova eina de codi obert capaç de resoldre els reptes en l'elaboració de perfils de matrius complexes).
RMN es una plataforma analítica utilizada para cuantificar los metabolitos presentes en las muestras de metabolómica. Los espectros de 1H-RMN muestran múltiples señales de metabolitos con tres parámetros específicos (desplazamiento químico, ancho medio de banda, intensidad) que pueden mostrar reactividad a las condiciones de la muestra. Esta reactividad perjudica a la optimización del fitado de los espectros necesario para realizar el perfilado automático de metabolitos de las muestras. El objetivo de esta tesis fue la exploración del uso de técnicas de tendencia basadas en Machine Learning (ML) con el uso de flujos de trabajo robustos para modelar y explotar la información presente en los diferentes parámetros de señal durante el perfilado de metabolitos de los conjuntos de datos 1H-NMR. En particular, las aplicaciones consideradas fueron la mejora de la clasificación de las muestras en los estudios de metabolómica y la mejora de la calidad del perfilado automático. Además de lograr estos objetivos, también se obtuvieron logros adicionales (por ejemplo, la generación de una nueva herramienta de código abierto capaz de resolver los retos en la elaboración de perfiles de matrices complejas).
NMR is an analytical platform used to quantify the metabolites present in metabolomics samples. 1H-NMR spectra show multiple metabolite signals, each one with three parameters (chemical shift, half bandwidth, intensity) which can show reactivity to the sample conditions. This reactivity is a challenge for the optimization of the lineshape fitting of spectra necessary to perform the automatic metabolite profiling of samples. The aim of this PhD thesis was the exploration of the use of trending machine learning (ML)-based techniques and of robust ML-based workflows to model and then exploit the information present in the different parameters collected for each signal during the metabolite profiling of 1H-NMR datasets. In particular, the applications considered were the enhanced classification of samples in metabolomics studies and the enhancement of the quality of automatic profiling in 1H-NMR datasets. in addition to the achievement of these goals, additional achievements (e.g., the generation of a new open-source tool able to solve challenges in the profiling of complex matrices) was also fulfilled.

Dans un marché globalisé où 40% du vin consommé est importé, le contrôle de la traçabilité est un enjeu majeur de la filière viti-vinicole. L'authentification du vin est le processus pouvant faire appel à différentes méthodes analytiques devant pouvoir contrôler trois paramètres fondamentaux : l’origine géographique, le cépage et le millésime. La spectroscopie RMN 1H quantitative (RMNq) est aujourd'hui considérée comme un outil très prometteur pour l'étude de l'authenticité des vins. Lors de cette thèse, une technique de dosage de 40 composés majoritaires des vins par RMN 1H a été développée. Elle permet l’acquisition d’une information riche et complexe en une seule analyse non-spécifique. La capacité de cette technique à authentifier un vin a été démontrée suite à une collaboration avec le Château Mouton-Rothschild, par comparaisons avec des analyses officielles réalisées par la DGDDI et DGCCRF de Pessac (SCL). La détermination d’un seuil de conformité a été établie en prenant en compte l’évolution naturelle des vins en bouteille. Une étude de caractérisation des vins rouges de Bordeaux a été menée. La singularité de ces vins a été observée par comparaison avec d’autres vins français, mettant en évidence des métabolites caractéristiques des vins de Bordeaux. Les résultats fournissent une description globale du potentiel de la RMN 1H pour l’authentification des vins
In a globalized market where 40% of the wine consumed is imported, traceability control is a major challenge for the wine industry. Wine authentication is the process that can use different analytical methods able to control three fundamental parameters: geographical origin, grape variety and vintage. Quantitative 1H NMR spectroscopy (qNMR) is now considered as a very promising tool for studying wine authenticity. During this thesis, a technique was developed for the determination of 40 major wine compounds by 1H NMR. It allows the acquisition of rich and complex information in a single non-specific analysis. The ability of this technique to authenticate a wine has been demonstrated following a collaboration with Château Mouton-Rothschild, by comparison with official analyses carried out by the DGDDI and DGCCRF of Pessac (SCL).The determination of a compliance threshold has been established by taking into account the natural evolution of bottled wines. A characterization study of Bordeaux red wines was carried out. The singularity of these wines was observed in comparison with other French wines, highlighting the characteristic metabolites of Bordeaux wines. The results provide a global description of the potential of 1H NMR for wine authentication

El perfilat de metabòlits es la tasca més difícil dins l'anàlisi espectral de RMN. El seu objectiu es comprendre els processos biològics que tenen lloc en un moment concret mitjançant la identificació i quantificació dels metabòlits presents en mescles d' RMN complexes. Un espectre de RMN està compost per ressonàncies d'un gran nombre de metabòlits, i aquestes se solen solapar entre elles, canviar de posició depenent del pH de la mostra i poden quedar emmascarades per senyals de macromolècules. Tots aquests problemes compliquen la identificació i quantificació de metabòlits, pel que obtenir un perfil de metabòlits curat en una mostra pot ser un gran repte inclús per usuaris experts. En aquest context, la motivació d'aquesta tesi va néixer amb l'objectiu de donar automatismes i funcions fàcils de fer servir per al perfilat de metabòlits en RMN, millorant la qualitat dels resultats i reduint el temps d'anàlisi. Per fer-ho, es van implementar un conjunt d'algoritmes que van acabar empaquetats en dos programes, Dolphin i Whale.
El perfilado de metabolitos es la tarea más difícil dentro del análisis espectral de RMN. Su objetivo es comprender los procesos biológicos que tienen lugar en un momento concreto a través de la identificación y cuantificación de los metabolitos presentes en mezclas de RMN complejas. Un espectro de RMN está compuesto por resonancias de un gran numero de metabolitos, y éstas a menudo se solapan entre ellas, cambian de posición dependiendo del pH de la muestra y pueden quedar enmascaradas por señales de macromoléculas. Todos estos problemas complican la identificación y cuantificación de metabolitos, por lo que obtener un perfilado de metabolitos curado en una muestra puede ser un gran reto incluso para usuarios expertos. En este contexto, la motivación de esta tesis nació con el objetivo de dar automatismos y funciones fáciles de usar para el perfilado de metabolitos en RMN, mejorando la calidad de los resultados y reduciendo el tiempo de análisis. Para hacerlo, se implementaron un conjunto de algoritmos que acabaron empaquetados en dos programas, Dolphin y Whale.
Metabolite profiling is the most challenging approach in NMR spectral analysis. It aims to comprehend biological processes occurring in a certain moment through identifying and quantifying metabolites present in complex NMR mixtures. An NMR spectrum is composed by resonances of a huge number of metabolites, and these resonances often overlap between them, shift position depending on the sample pH and can be masked by macromolecules signals. All these drawbacks hinder metabolite identification and quantification, so obtaining a cured metabolite profile of a sample can be a very big issue even for expert users. In this context, the motivation of this thesis was born with the aim to provide automatisms and user-friendly interactive functions for NMR metabolite profiling, improving the quality of the results and reducing the time span of the analysis. To do so, several algorisms were implemented and embedded into two software packages, Dolphin and Whale.

Metabolomics seeks to achieve a comprehensive quantitative analysis of the wide range of metabolites in biological samples. Analytical chemistry, and specifically 1H NMR, can provide spectra detailing the vast array of physio-chemical properties of the sample. The identification of characteristics in the spectra is known as metabolomic fingerprinting. However, the thousands of variables in the 1H NMR spectra, .relating to hundreds of metabolites, make the data sets extremely complex and difficult to interpret. The aim of this research was to develop analytical techniques to classify biological samples, which are able to identify the metabolites responsible for the differences between classes. Novel methods based on genetic programming have been developed which provide models that are able to classify samples at least as well as currently used chemometric pattern recognition approaches. Moreover these techniques have the advantage that they are significantly easier to interpret in terms of the original spectrum. Preprocessing the spectra using wavelet transforms has further increased the interpretability of the models. This adaptive binning method integrates the data in such a way that the bins represent peaks in the original spectra thereby reducing the dimensionality whilst maintaining the information. The preprocessing and genetic programming methods have been combined and developed in response to additio~al problems faced by time resolved metabolomics data, where both intra-individual (time dependent) and inter-individual variation are present. The methods were employed in the analysis of TSE infected sheep and cattle and, in both cases, enabled a diagnostic of the disease to be determined, allowing the specific compounds responsible for the disease to be identified.

Background Prostate cancer (PCa) is the most frequently diagnosed malignant disease among adult males in the USA and the second leading cause of cancer deaths in men. Due to the lack of diagnostic tools that are able to differentiate highly malignant and aggressive cases from indolent tumors, overtreatment has become very common in the era of prostate specific antigen (PSA) screening. New diagnostic methods to determine biological status, malignancy, aggressiveness and extent of PCa are urgently needed. 1H High Resolution Magic Angle Spinning Nuclear Magnetic Resonance Spectroscopy (1H HRMAS MRS) can be used to establish PCa metabolomic profiles while preserving tissue architecture for subsequent histopathological analysis. Immunohistochemistry (IHC), as opposed to conventional histopathology methods, has the potential to provide objective, more accurate and quantitative knowledge of tissue pathology. This diagnostic- accuracy study sought to evaluate a novel approach to quantitatively identify metabolomic markers of PCa by exploring the potential of PCa immunomarkers to quantify metabolomic profiles established by 1H HRMAS MRS. Material and Methods 1H HRMAS MRS was performed on tissue samples of 51 prostate cancer patients using a 14.1 Tesla NMR spectrometer (BRUKER Biospin, Billerica, MA) with a rotor synchronized CPMG pulse sequence. Spectral intensities of 36 regions of interest were measured as integrals of curve fittings with Lorentzian-Gaussian line shapes. Immunohistochemistry (IHC) was carried out following the spectroscopy scan, using three prostate immunomarkers to identify cancerous and benign glands: P504S (Alpha-methylacyl-CoA-racemace), CK903 (high-molecular weight cytokeratin) and p63. The immunostaining quality following 1H HRMAS MRS was evaluated and compared to unscanned sections of the same sample, to verify the stability and accessibility of the proposed immunomarkers. IHC images were automatically and quantitatively evaluated, using a quantitative image analysis program (QIAP), to determine the percentage of cancerous and benign epithelia in the tissue cross- sections. The results of the program were validated by a correlation with the results of a quantitative IHC review and quantitative conventional histopathology analysis performed by an experienced pathologist. Ultimately, spectral intensities and the cancer epithelium percentage, obtained from quantitative immunohistochemistry, were correlated in order to validate PCa metabolomic markers identified by 1H HRMAS MRS. Patient outcomes and incidence of recurrence were determined by retrospective review of medical records five years after initial surgery. Categories of recurrence were correlated to spectral intensities to explore potential metabolomic markers of recurrence in the cohort. Results Immunostainings with P504S and CK903 showed excellent staining quality and accessibility following 1H HRMAS MRS, suggesting these markers to be suitable for the presented quantitative approach to determine metabolomics profiles of PCa. In contrast, the quality of p63 IHC was impaired after previously performed spectroscopy. IHC using the immunomarkers P504S and CK903 on adjacent slides was found to present a feasible quantitative diagnostic method to distinguish between benign and cancerous conditions in prostate tissue. The cancer epithelium percentage as determined by QIAP showed a significant correlation to the results of quantitative IHC analysis performed by a pathologist (p < 0.001), as well as to a quantitative conventional histopathology review (p = 0.001). The same was true for the benign epithelium percentage (p < 0.001 and p = 0.0183), validating the presented approach. Two metabolomic regions showed a significant correlation between relative spectral intensities and the cancer epithelium percentage as determined by QIAP: 3.22 ppm (p = 0.015) and 2.68 ppm (p = 0.0144). The metabolites corresponding to these regions, phosphocholine and citrate, could be identified as metabolomic markers of PCa in the present cohort. 45 patients were followed for more than 12 months. Of these, 97.8% were still alive five years after initial surgery. 11 patients (24.4%) experienced a recurrence during the follow- up time. The categories of recurrence showed a correlation to the spectral intensities of two regions, 2.33 – 2.3 ppm (p = 0.0403) and 1.28 ppm (p = 0.0144), corresponding to the metabolites phosphocreatine and lipids. Conclusion This study introduces a method that allows an observer-independent, quantitative analysis of IHC to help establish metabolomic profiles and identify metabolomic markers of PCa from spectral intensities obtained with 1H HRMAS NMR Spectroscopy. The immunomarkers P504S and CK903 have been found suitable IHC analysis following 1H HRMAS MRS. A prospective in vivo application of PCa metabolite profiles and metabolomic markers determined by the presented method could serve as highly sensitive, non- invasive diagnostic tool. This observer- independent, computer- automated, quantitative analysis could help to distinguish highly aggressive tumors from low-malignant conditions, avoid overtreatment and reduce risks and complications for cancer patients in the future. Further studies are needed to verify the identified PCa metabolomic markers and to establish clinical applicability
Einführung Prostatakrebs ist eine häufigsten Krebserkrankungen in den USA und die zweithäufigste malignom- assoziierte Todesursache männlicher Patienten weltweit. Seit der Einführung des Prostata- spezifischen Antigen (PSA)- Screeningtests wird diese Krebsart in früheren Stadien diagnostiziert und therapiert, wodurch die Mortalitätsrate in den letzten Jahren deutlich reduziert werden konnte. Da moderne diagnostische Methoden bislang jedoch nicht ausreichend in der Lage sind, suffizient zwischen hochmalignen und weniger aggressiven Varianten dieses bösartigen Krebsleidens zu unterscheiden, werden häufig auch Patienten aggressiv therapiert, deren niedriggradiges Prostatakarzinom keine klinische Relevanz gehabt hätte. Es besteht daher ein großes wissenschaftliches Interesse an der Entwicklung neuer diagnostischer Methoden zur akkuraten Bestimmung von biologischem Status, Malignität, Aggressivität und Ausmaß einer Prostatakrebserkrankung. \\\\\\\"1H High Resolution Magic Angle Spinning Nuclear Magnetic Resonance Spectroscopy\\\\\\\" (1H HRMAS MRS) ist eine vielversprechende diagnostische Methode, welche es ermöglicht, metabolomische Profile von Prostatakrebs zu erstellen, ohne die Gewebsstruktur der analysierten Proben zu zerstören. Durch anschließende histopathologische Begutachtung lassen sich die erstellten Metabolitprofile validieren und evaluieren. Im Gegensatz zu konventionellen histopathologischen Methoden können durch immunhistochemische Verfahren dabei objektivere, akkuratere und quantifizierbare histopathologische Erkenntnisse gewonnen werden. Die vorliegende Studie präsentiert einen neuentwickelten diagnostischen Ansatz zur quantitativen Bestimmung von metabolomischen Markern von Prostatakrebs, basierend auf der Durchführung von 1H HRMAS NMR Spektroskopie und quantitativer Immunhistochemie. Material und Methoden Einundfünfzig Gewebsproben von Prostatakrebspatienten wurden mittels 1H HRMAS MRS an einem 14.1 T BRUKER NMR Spektrometer unter Einsatz einer CPMG-Pulssequenz untersucht. Spektrale Intensitäten in 36 Metabolitregionen wurden gemessen. Anschließend wurden die analysierten Gewebeproben mit drei Immunfärbemarkern für sowohl malignes (P504S, Alpha-methylacyl-CoA-racemase) als auch benignes (CK903, High-molecular weight cytokeratin, und p63) Prostatagewebe angefärbt und quantitativ mit Hilfe eines Bildanalyseprogramms (QIAP) ausgewertet. Die Anwendbarkeit und Auswertbarkeit der genannten Immunomarker nach Spektroskopie wurde evaluiert und mit der Färbungsqualität von nicht- gescannten Schnitten verglichen. Die Resultate der automatischen Auswertung durch QIAP konnten durch einen erfahrenen Pathologen in einer quantitativen Analyse der Immunfärbungen sowie konventioneller histologischer Färbungen derselben Gewebsproben validiert werden. Die spektralen Intensitäten aus den Messungen mit 1H HRMAS MRS wurden mit den korrespondierenden Ergebnissen der quantitativen Auswertung der Immunfärbungen korreliert, um metabolomische Marker von Prostatakrebs zu identifizieren. Der klinische Verlauf und die Rezidivrate der Patienten wurden 5 Jahre nach der initialen Prostatektomie retrospektiv bestimmt. Rezidivkategorien wurden erstellt und mit den bestimmten spektralen Intensitäten korreliert, um metabolomische Marker für das Auftreten von Prostatakrebsrezidiven zu identifizieren. Ergebnisse Die Immunfärbungen mit P504S und CK903 zeigten exzellente Qualität und Auswertbarkeit nach vorheriger 1H HRMAS MRS. Beide Marker eigneten sich zur Durchführung von quantitativer Immunhistochemie an spektroskopierten Gewebeproben. Im Gegensatz dazu war die Qualität der Immunfärbungen mit p63 nach Spektroskopie vermindert. Quantitative Immunfärbungen unter Einsatz der Immunmarker P504S und CK903 stellten eine praktikable diagnostische Methode dar, um zwischen malignen und benignem Prostatagewebe zu unterscheiden. Der Anteil von bösartig verändertem Prostatagewebe, bestimmt durch QIAP, korrelierte signifikant mit den Ergebnissen der quantitativen Analyse der Immunfärbungen durch den Pathologen (p < 0.001), sowie mit der quantitativen Auswertung der konventionellen histopathologischen Färbung (p = 0.001). Ebenso ließ sich die Bestimmung des Anteils von benignem Gewebe mit QIAP zu den Ergebnissen der pathologischen Analyse korrelieren (p < 0.001 und p = 0.0183). Für zwei metabolomische Regionen konnte ein signifikante Korrelation zwischen relativen spektralen Intensitäten, bestimmt mit 1H HRMAS NMR Spektroskopie, und dem Anteil von malignem Epithelium in derselben Gewebeprobe, ermittelt durch QIAP, festgestellt werden: 3.22 ppm (p = 0.015) und 2.68 ppm (p = 0.0144). Die zu diesen Regionen korrespondierenden Metaboliten, Phosphocholin und Zitrat, konnten als potentielle metabolomische Marker für Prostatakrebs identifiziert werden. Die retrospektiven Analyse der klinischen Daten der Patienten fünf Jahre nach Prostatektomie ergab eine Überlebensrate von 97.8%. Elf dieser Patienten (24.4%) erlitten ein Rezidiv ihrer Erkrankung. Die bestimmten Rezidivkategorien korrelierten signifikant mit zwei metabolomischen Regionen (2.33 – 2.3 ppm, p = 0.0403 und 1.28 ppm, p = 0.0144), welche zu den Metaboliten Phosphokreatin und Lipiden korrespondierten. Schlussfolgerung Die vorliegende Studie präsentiert einen diagnostischen Ansatz zur objektiven und quantitativen Bestimmung metabolomischer Marker von Prostatakrebs unter Verwendung von 1H HRMAS MRS und Immunhistochemie. P504S und CK903 eignen sich als Immunmarker für quantitative Immunfärbungen nach vorheriger Durchführung von 1H HRMAS MRS. Die Metaboliten Phosphocholin und Zitrat konnten in der vorliegenden Patientenkohorte als potentielle metabolomische Marker für Prostatakrebs identifiziert werden. Eine mögliche in vivo Anwendung der gefundenen metabolomischen Marker könnte als hochsensitives, objektives und nicht- invasives diagnostisches Werkzeug der Prostatakrebsdiagnostik dienen. Der vorliegende untersucherunabhängige, automatisierte und quantitative diagnostischer Ansatz hat das Potential, zwischen hochmalignen und weniger aggressiven Krebsfällen zu unterscheiden und somit unnötige Risiken und Komplikationen für Prostatakrebspatienten zu reduzieren. Weitere Untersuchungen sind notwendig, um die identifizierten metabolomischen Marker zu verifizieren und eine klinische Anwendung zu etablieren

L'attenzione della comunità sportiva verso l'uso di integratori a base di probiotici come mezzo per migliorare le prestazioni fisiche, oltre che la buona salute, sono aumentate negli ultimi anni. Questo è vero anche per i cavalli, con promettenti risultati. In questo lavoro di tesi è descritta una serie di test su un mix probiotico con diversi ceppi di batteri vivi tipicamente impiegati per l'uomo, usato al fine di migliorare le prestazioni in allenamento di cavalli da trotto. Per valutare gli effetti del mix sulla performance dei cavalli, è stata misurata la concentrazione di lattato nel sangue, un dato traslazionale frequentemente impiegato allo scopo, combinato con lo studio del metaboloma urinario. I risultati suggeriscono che la supplementazione con i probiotici testati ha ridotto significativamente la concentrazione di lattato nel sangue post-esercizio. Le modifhce notate nel metaboloma urinario, ha suggerito che un probabile meccanismo alla base di questo effetto positivo era collegato ad uno switch della fonte di energia nel muscolo dai carboidrati agli acidi grassi a catena corta. Tre molecole contenenti zolfo diversamente concentrate nelle urine in relazione ai probiotici ha suggerito che tale cambiamento sia collegato al metabolismo dello zolfo.

Metabolomics is a technique that can be used to investigate the metabolic profiles of an organism by measuring a large number of the low molecular weight metabolites in the metabolic pool. 1H NMR spectroscopy is an analytical tool that is unbiased, and a non-destructive way of investigating metabolic perturbations in organisms exposed to environmental stress. In this study two model organisms, Mytilus edulis and Chironomus riparius were used to investigate the effects of several pesticides: lindane and atrazine in the mussels, and fenitrothion, methiocarb and permethrin in the midge larvae. The impact of hypoxia and starvation were also investigated alongside the exposure to atrazine in the mussel. The pesticides used in the mussel study have different modes of action, but produce similar changes to behaviour and can cause starvation and mild hypoxia. Acetonitrile/²H₂O (60/40 % v/v) extracts of foot muscle of mussels subjected to hypoxia, or starvation, or to low or high doses of pesticide were analysed using ¹H NMR spectroscopy to produce metabolic profiles associated with these treatments. Discriminant analysis showed significant differences between treated and control animals and gave a clear separation between all treatment groups. Atrazine profiles were clearly separated from the starved and hypoxic animals and the animals exposed to high and low doses were also separated. Lindane treatment was separated from control animals in a dose-dependent way. This was associated with an increase in alanine concentrations and a decrease in all other identified metabolites. The study of midge larvae used the same approach, but using extracts of pooled whole body homogenates (10 animals per sample) instead of tissue from individuals. The animals were subjected to low, environmentally relevant dose levels of three pesticides; fenitrothion, methiocarb and permethrin. The first two of these have a similar mode of action, inhibition of acetylcholine esterase, whilst the latter is an axonal poison acting on cation channels. The metabolic profiles associated with these treatments showed a clear separation between all treatment groups and between treated and control animals. Fenitrothion treatment was associated with an increase in alanine (on average of 93.3 M relative to controls) and lactate concentrations compared with controls and other treatments. Methiocarb caused a reduction in arginine, leucine and lysine concentrations to half of the control level. Permethrin produced a reduction in tyrosine and phenylalanine concentrations to half that of the control group. In both these experiments the use of ¹H NMR metabolomics enabled the separation the effects of all of the treatments and stressors from each other and from the controls. It demonstrates the potential of the metabolomic approach to provide separation of the effects of poisoning from those of environmental stress, and to distinguish between toxicants with similar modes of action.

1H nuclear magnetic resonance (NMR) metabolomics was used to determine the response of earthworm exposure to polycyclic aromatic hydrocarbons (PAHs) in contact and soil tests. Eisenia fetida is recommended for toxicology testing, but to date this species has not frequently been used in environmental metabolomic studies. The metabolic profile of E. fetida was characterized with the goal of using this species in metabolomic studies. Testing several individual solvents for earthworm tissue extraction indicated that D2O buffer extracted the highest concentration of the widest variety of earthworm metabolites. Sample preparation methods were evaluated to reduce variability and achieve reproducible control groups for use in metabolomic studies. 96h depuration and intact lyophilization of earthworms before homogenization resulted in the least variation between sample extracts. This sample preparation method was used to compare E. fetida and two other earthworm species (Lumbricus rubellus and Lumbricus terrestris) and E. fetida had the most reproducible 1H NMR spectra. E. fetida was then used to identify the metabolic response after exposure to several concentrations of the polycyclic aromatic hydrocarbons (PAHs) naphthalene, phenanthrene and pyrene, individually and in mixtures. With exposure to individual PAHs in contact tests, earthworm responses were both PAH- and concentration- dependent. In earthworms exposed to PAH mixtures in contact tests, an increase in amino acids was measured. Furthermore, an increase in specific amino acids and a decrease in maltose were identified as potential indicators of sub-lethal phenanthrene exposure in soil. Lastly, the relationship between earthworm response and contaminant bioavailability in soil was tested. Contaminant bioavailability is typically assessed using indirect methods [e.g., ‘soft’ extraction techniques like hydroxypropyl cyclodextrin (HPCD) extraction]. However, it was found that the directly measured response of earthworm exposure to sub-lethal concentrations of phenanthrene in soil is related to both the total and bioavailable phenanthrene. This suggests there is potential for the use of 1H NMR metabolomics for the assessment of contaminant bioavailability. This thesis has demonstrated that E. fetida are suitable for metabolomic studies and has indicated that 1H NMR metabolomics may have potential for measuring and monitoring earthworm exposure to sub-lethal concentrations of PAHs.

碩士
國立臺灣大學
環境衛生研究所
100
Naphthalene, a primary polycyclic aromatic hydrocarbon (PAH), widely spread in the environment was hypothesized to relate with adverse health effects. In order to clarify the underlying mechanisms of naphthalene-induce toxicity, extensive studies are required. Therefore, this study intended to investigate the mechanisms of naphthalene-induce toxicity in various tissues of mice. We used high through-put 1H NMR-based metabolomics to study effects of naphthalene on mouse tissues. Dose-response experiments (0, 100 and 200 mg/kg naphthalene, ip) were carried out on male ICR mice. Both hydrophilic and hydrophobic metabolites from the lung, liver and kidney tissues were extracted and analyzed by 1H and p-J-resolved NMR followed by principal component analysis (PCA) and orthogonal projections to latent structures discriminant analysis (OPLS-DA). From our results, the metabolic effects of naphthalene on lung, liver, and kidney demonstrated dose-dependent effects. Our results suggest that naphthalene-induced lung injures were associated with cellular membrane damages and turbulence of energy metabolisms, where declination of glycerophosphocholine and increase of succinate was found in the lung. In the liver, glutamine and UDP-glucose were increased in the 200 mg/kg exposure group. Besides, detoxification nucleophile glutathione, also demonstrated to be increased depending on naphthalene dose in the kidney. Through this study, not only the metabolome of target organ lung but metabolome of liver and kidney were found to respond to naphthalene intervention. The changes of these metabolites may partially explain the pathological response of naphthalene exposure. In conclusion, 1H NMR-based metabolomics is a useful tool in understanding the mechanism of xenobiotic induced toxicities.

碩士
國立臺灣大學
環境衛生研究所
100
Although epidemiological studies have demonstrated adverse human health effects with a greater risk for cancer development when subjects are exposed to ionizing radiation, the mechanisms of biological impacts are still unclear despite the increasing use of diagnostic radiology in medicine. In recent years, several functional genomic approaches have been developed and applied to examine molecular events in biological systems exposed to ionizing radiation. Little metabolomic studies were conducted on the metabolic effects of ionizing radiation. Therefore, we intend to use metabolomic approach to understand the molecular events in a more functional measurement. Human lymphoblastic cell lines: TK6 (wild-type p53) and WTK1 (mutant p53) were used to examine metabolic effects of ionizing radiation. Time-course and dose-response experiments were conducted in cells treated with gamma-ray （iso-survival dose, D0 (TK6: 0.8 Gy; WTK1: 1.5 Gy） or 10 Gy) for 3 or 24 hours. Hydrophilic and hydrophobic metabolites were extracted and analyzed by 1H and J-resolved NMR followed by principal component analysis and metabolite identification. Based on our results, both hydrophilic and hydrophobic metabolome are shown different radiation effects between TK6 and WTK1 cells. The γ-ray radiation effects on hydrophilic metabolome at 24 hours are greater than at 3 hours both in different cell types and doses. The dose effect of ionizing radiation on the WTK1 may be related to intracellular oxidative stress. In summary, after γ-ray exposure, the hydrophilic metabolome changing in cells may caused by oxidative stress, cell membrane damage, apoptosis and DNA damage. The various p53 status also results in differences in taurine, alanine, creatine and glutathione between TK6 and WTK1 cells.

碩士
國立臺灣大學
環境衛生研究所
101
Naphthalene which is widely spread in the environmental is a common polycyclic aromatic hydrocarbon. Several studies indicated that single dose exposure of naphthalene cause acute injury on mouse Clara cell, a bronchiolar epithelial cell type. However, when mice are administered repeated doses of naphthalene, the susceptible Clara cell become refractory to injury. In this study, we intended to investigate the mechanisms of naphthalene toxicity in various mouse tissues among injured, tolerant, and the control mice using 1H NMR-based metabolomics. Male ICR mice were administered seven repeated injections (ip) of NA (0, 200 mg/kg/day; single and repeated naphthalene exposure, respectively) in olive oil and gave a challenge dose (300 mg/kg/day) at eighth day. Bronchoalveolar lavage fluid and both hydrophilic and hydrophobic extracts from the lung, liver and kidney were analysed by 1H NMR followed by principal component analysis. Our results showed that, the metabolites effect of single naphthalene which cause cell injury on respiratory system are associated with cellular membrane damages and energy metabolism disturbance. However, the repeated exposure may induce the antioxidation mechanism associated with glutathione; Therefore mice become tolerant to naphthalene. In addition, we also suspected that repeated exposed to naphthalene would cause lung tumor development, since declination of glucose and increase of glutamine was found in the lung of repeated exposure group. In the liver, the upregulation of glutathione in repeated exposure mice may play an important role in leading to naphthalene toxicity tolerance, too. The metabolome disturbance of single exposure in the liver and kidney are associated with energy metabolism. In this study, we have partially explained the mechanisms of naphthalene toxicity and detoxification within single dose and repeated exposure. In conclusion, 1H NMR-based metabolomics is useful in understanding the toxicity and detoxifying mechanisms due to xenobiotic interventions.

碩士
國立臺灣大學
環境衛生研究所
104
More and more food safety issues are noticed by the public. In 2013, the Ministry of Health and Welfare (MOHW) in Taiwan declared that some starch-processed foods were illegally added food addictive, maleic acid/maleic anhydride. Modified starch can enhance favorable properties, such as viscosity, texture, and elasticity in food. Accidental consumption of maleic acid at low levels does not cause significant adverse health effects; however, long term exposure of high levels of maleic acid can induce kidney damage. The molecular effects of repeated maleic acid exposure are still largely unknown. In this study, we intend to understand metabolic effects of repeated exposure to maleic acid in rats using 1H NMR-based metabolomic approach. Rat urinary metabolome were examined to study time-course and dose-response of maleic acid. Adult male Sprague-Dawley (SD) rats were divided into control, low-dose (6 mg/kg), medium-dose (20 mg/kg), and high-dose (60 mg/kg) and treated with vehicle or maleic acid via oral gavage daily. Urine samples were collected twice a day (once during daytime and once at night) on day 0, 7, 14, 21, and 28 and then examined by high-resolution 1H nuclear magnetic resonance (NMR) followed by multivariate statistical analysis. The principle component analysis (PCA) score plots from the anlaysis of urinalry metabolome showed changes of metabolome patterns within different exposure groups. Clear metabolome seperation between high-dose and the control groups were observed from the night samples of day 14 and later. The increased levels of acetoacetate and hippurate, and decreased levels of alanine and acetate in the treatment groups were observed in the night samples of day 28. Changes of metabolites are related with environment stress and energy metabolism. Metabolic effects of maleic acid exposure are obvious on rats in high dose group and at last time points. By investigating the perturbation of urinary metabolome in SD rats can assist urinary biomarker discovery for maleic acid and find out possible toxic mechanisms induced by maleic acid.

Chapter 1 is divided in 4 parts, comprising a first part referring to the definition of preterm, identifying its main associated risk factors and consequent health complications in short and long terms. In part 2 the applied methodology is presented, giving a brief introduction about the analytical and statistical methods used. The principles of the methodology used and state of the art are also described. Part 3 describes the importance of biomarkers in diagnostics, therapy and prognostics of disease. In part 4 the clinical potential of metabolomics in neonatology is described, referring the biofluids used and the main metabolic studies associated with preterm birth (PTB). This part concludes with a literature review of studies of premature infants and newborns, and premature infants compared to other disorders. Chapter 2 describes the experimental procedures used to perform this work, including group presentation, sample collection and preparation, sample acquisition and data processing. Chapter 3 characterizes the composition of newborn’s urine using Nuclear Magnetic Resonance (NMR) spectroscopy, which 49 metabolites were identified. Chapter 4 presents the metabolic study of preterm newborns to find markers related to prematurity at the time of birth. Metabolic variations at different stages of prematurity (extreme preterm (<28gestational weeks), very preterm (28<32 gestational weeks) and moderate to late preterm (32<37gestational weeks)) are presented. Chapter 5 deals with the metabolomic study of the development of preterm newborns over time, during their stay in hospital. Finally, the main conclusions of this work are present in Chapter 6, emphasizing the clinical potential of metabolomics in preterm urine for monitoring newborns during their hospital stay, in order to use new biomarkers of deviant behaviours indicative of health complications during that period.
O capítulo 1 divide-se em quatro partes, sendo a primeira parte referente à definição de parto prematuro, identificando os seus principais fatores de riscos e consequentes complicações na saúde a curto e longo prazo. Na parte 2 é apresentada a metodologia aplicada, fazendo-se uma breve introdução aos métodos analíticos e estatísticos usados e sendo ainda referidos os princípios da metodologia usada e estado da arte. A parte 3 descreve a importância dos biomarcadores no diagnóstico, terapia e prognóstico na área da saúde, referindo como são identificados e validados clinicamente. Na parte 4 é descrito o potencial clínico da metabolómica na neonatologia, fazendo referência aos biofluidos usados e aos principais estudos metabolómicos associados ao parto prematuro. Esta parte termina com uma revisão da literatura de estudos realizados em prematuros recémnascidos. O capítulo 2 descreve os procedimentos experimentais utilizados para a realização deste trabalho, incluindo apresentação dos grupos, recolha e preparação de amostras, aquisição de amostras e tratamento de dados. No capítulo 3 é feita a caracterização da composição da urina de recém-nascidos através de espectroscopia de Ressonância Magnética Nuclear (RMN) através da qual foram identificados 49 metabolitos. O capítulo 4 apresenta o estudo metabolómico de recém-nascidos prematuros a fim de encontrar marcadores relacionados com prematuridade no momento do nascimento. Nesse contexto são apresentadas as variações metabólicas em diferentes estadios de prematuridade (extremo (<28semanas gestacionais), muito prematuro (28<32 semanas gestacionais) e moderado a tardio (32<37semanas gestacionais)). No capítulo 5 é feito o estudo metabolómico relativamente ao desenvolvimento dos recém-nascidos prematuros durante a sua estadia no hospital até ao momento da alta médica. Finalmente, no capítulo 6 apresentam-se as principais conclusões deste trabalho, enfatizando o potencial clínico da metabolómica de urina de recém-nascidos prematuros para o acompanhamento destes bebés durante a estadia no hospital, visando o uso de novos biomarcadores indicativos de desvios relativos a complicações de saúde durante esse período.
Mestrado em Bioquímica

碩士
國立臺灣大學
環境衛生研究所
103
In the manufactoring processes of synthetic leather, the workers might expose to variety of solvents, such as N,N-dimethylformamide (DMF), toluene (TOL), methyl ethyl ketone (MEK) which were all confirmed to have adverse health effects toward human by previous studies. This reasearch is to study metabolic effects of numerous co-exposure solvents on the workers of synthetic leather factories by using 1H nuclear magnetic resonance (NMR)-based metabolomic approach. Biological samples, urine, were collected from workers (n=39) of a synthetic leather factory in one day at 2 time points, pre-shift and post-shift. Metabolic profile of each sample was analyzed by 500 MHz NMR. Air samples were also collected using personal air-sampling to evaluate ambient concentration of DMF, TOL, MEK, and other solvents. Principal components analysis (PCA), partial least‐squares discriminant analysis (PLS‐DA), and logistic regression analysis were used to assess the association between metabolites and solvents. The results indicated that decreased level of 2-oxoisovalerate and elevated level of creatinine in pre-shift urine were associated with long-term exposure of higher level of DMF or higher level of TOL. Also, exposure to high level of TOL caused elevated level of hippurate and exposure to high level of MEK down regulated urinary histidine level. In addition, exposure to DMF and TOL simultaneously could enhance the effect on metabolic profiles that induced increased level of urinary 2-oxisovalerate, hippurate, 3-hydroxybutyrate and 3-aminoisobutyrate; depletion of guanidoacetate and phenylalanine. Overall, we discovered interference of the energy metabolism and amino acids metabolism when workers experienced multiple solvents exposure in the synthetic leather factory. Moreover, exposure to DMF and TOL simultaneously should be avoided by the workers and appropriate personal protective equipment is necessary to diminish worker’s exposure level. With the knowledge of this research, we can discover new biomarkers to identify adverse health effects induced by the solvents and conduct a better exposure assessment in order to give workers a better and safer occupational environment.

There is a growing need to develop rapid and cost-effective ecotoxicological tools for risk assessment because traditional methods examine endpoints such as mortality, which do not provide any insight into the mode of action (MOA) of the chemical. Research presented within this thesis illustrates the potential of 1H NMR-based metabolomics as a rapid and routine ecotoxicological tool that can elucidate a chemical’s MOA and also aid in the identification of metabolites of exposure. Metabolomics involves measuring the fluctuations in the endogenous metabolites of an organism within a cell, tissue, bio-fluid or whole organism in response to an external stressor. We focused on the model polycyclic aromatic hydrocarbon (PAH) phenanthrene, and the perfluoroalkyl acids (PFAAs) perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS), due to their recalcitrant nature and widespread prevalence in soil environments. 1H NMR-based metabolomics analysis of the exposure of Eisenia fetida earthworms to sub-lethal phenanthrene exposure via filter paper contact tests revealed a concentration-dependent two-phased MOA: a linear correlation between the metabolic response and exposure concentration at low concentrations followed by a plateau in the responses at high concentrations. Alanine, glutamate, maltose, cholesterol and phosphatidylcholine emerged as potential indicators of phenanthrene exposure. An increased energy demand and an interruption in the conversion of succinate to fumarate in the Krebs cycle were observed due to phenanthrene exposure. Sub-lethal PFOA and PFOS exposure to E. fetida via contact tests for two days revealed heightened responses with higher PFOA and PFOS concentrations. Leucine, arginine, glutamate, maltose, and ATP were identified as potential indicators of PFOA or PFOS exposure. E. fetida responses were then investigated after exposure for two, seven and fourteen days to an artificial soil that was spiked with sub-lethal PFOS concentrations. An exposure time-dependent operation of two separate MOAs were identified. Both the contact tests and artificial soil exposure studies identified an elevation in fatty acid oxidation, a disruption in energy metabolism and biological membrane structure, and also an interruption of ATP synthesis following PFOA and PFOS exposure. This thesis illustrates the promise of NMR-based metabolomics as a routine tool for ecotoxicological assessment of contaminated sites.

Asthma is a significant health issue in the pediatric population with a noteworthy growth over the years. The proposed challenge for this PhD thesis was the development of advanced methodologies to establish metabolomic patterns in urine and exhaled breath associated with asthma whose applicability was subsequently exploited to evaluate the disease state, the therapy adhesion and effect and for diagnostic purposes. The volatile composition of exhaled breath was studied combining headspace solid phase microextraction (HS-SPME) with gas chromatography coupled to mass spectrometry or with comprehensive two-dimensional gas chromatography coupled to mass spectrometry with a high resolution time of flight analyzer (GC×GC–ToFMS). These methodologies allowed the identification of several hundred compounds from different chemical families. Multivariate analysis (MVA) led to the conclusion that the metabolomic profile of asthma individuals is characterized by higher levels of compounds associated with lipid peroxidation, possibly linked to oxidative stress and inflammation (alkanes and aldehydes) known to play an important role in asthma. For future applications in clinical settings a set of nine compounds was defined and the clinical applicability was proven in monitoring the disease status and in the evaluation of the effect and / or adherence to therapy. The global volatile metabolome of urine was also explored using an HSSPME/GC×GC–ToFMS method and c.a. 200 compounds were identified. A targeted analysis was performed, with 78 compounds related with lipid peroxidation and consequently to oxidative stress levels and inflammation. The urinary non-volatile metabolomic pattern of asthma was established using proton nuclear magnetic resonance (1H NMR). This analysis allowed identifying central metabolic pathways such as oxidative stress, amino acid and lipid metabolism, gut microflora alterations, alterations in the tricarboxylic acid (TCA) cycle, histidine metabolism, lactic acidosis, and modification of free tyrosine residues after eosinophil stimulation. The obtained results allowed exploring and demonstrating the potential of analyzing the metabolomic profile of exhaled air and urine in asthma. Besides the successful development of analysis methodologies, it was possible to explore through exhaled air and urine biochemical pathways affected by asthma, observing complementarity between matrices, as well as, verify the clinical applicability.

碩士
國立高雄海洋科技大學
水產養殖研究所
104
There are two different ways in Artemia breeding strategies, oviparity and ovoviviparity. Nowadays, many studies had demonstrated that the diapause eggshell of Artemia is necessary for oviparity to successful survival in the extreme environments, but the discrepancy of mechanism between oviparous and ovoviviparous is still unclear. In order to reveal the differences, two ways of Artemia breeding strategies are analyzed through NMR techniques, Metabolomics. The result suggested that several metabolites including 5,6-dihydrouracil, betaine, malate, methylamine, methylguanidine, N-acetylaspartate, succinate, trimethylamine and β-alanine are significantly different between oviparous and ovoviviparous stages. The analysis of the Principal Component Analysis (PCA) and Partial Least Squares-Discriminant Analysis (PLS-DA) also strongly indicated that the differences between oviparity and ovoviviparity are significant. The consequences of Loading bi plot showed that the nine metabolites involved in ovoviviparity are very important. Taken together, Artemia under extreme environment may initiate different metabolism, including β-alanine and pyrimidine metabolism, pantothenate and co-A biosynthesis and citrate cycle (TCA cycle) these pathway.

Background Prostate cancer (PCa) is the most frequently diagnosed malignant disease among adult males in the USA and the second leading cause of cancer deaths in men. Due to the lack of diagnostic tools that are able to differentiate highly malignant and aggressive cases from indolent tumors, overtreatment has become very common in the era of prostate specific antigen (PSA) screening. New diagnostic methods to determine biological status, malignancy, aggressiveness and extent of PCa are urgently needed. 1H High Resolution Magic Angle Spinning Nuclear Magnetic Resonance Spectroscopy (1H HRMAS MRS) can be used to establish PCa metabolomic profiles while preserving tissue architecture for subsequent histopathological analysis. Immunohistochemistry (IHC), as opposed to conventional histopathology methods, has the potential to provide objective, more accurate and quantitative knowledge of tissue pathology. This diagnostic- accuracy study sought to evaluate a novel approach to quantitatively identify metabolomic markers of PCa by exploring the potential of PCa immunomarkers to quantify metabolomic profiles established by 1H HRMAS MRS. Material and Methods 1H HRMAS MRS was performed on tissue samples of 51 prostate cancer patients using a 14.1 Tesla NMR spectrometer (BRUKER Biospin, Billerica, MA) with a rotor synchronized CPMG pulse sequence. Spectral intensities of 36 regions of interest were measured as integrals of curve fittings with Lorentzian-Gaussian line shapes. Immunohistochemistry (IHC) was carried out following the spectroscopy scan, using three prostate immunomarkers to identify cancerous and benign glands: P504S (Alpha-methylacyl-CoA-racemace), CK903 (high-molecular weight cytokeratin) and p63. The immunostaining quality following 1H HRMAS MRS was evaluated and compared to unscanned sections of the same sample, to verify the stability and accessibility of the proposed immunomarkers. IHC images were automatically and quantitatively evaluated, using a quantitative image analysis program (QIAP), to determine the percentage of cancerous and benign epithelia in the tissue cross- sections. The results of the program were validated by a correlation with the results of a quantitative IHC review and quantitative conventional histopathology analysis performed by an experienced pathologist. Ultimately, spectral intensities and the cancer epithelium percentage, obtained from quantitative immunohistochemistry, were correlated in order to validate PCa metabolomic markers identified by 1H HRMAS MRS. Patient outcomes and incidence of recurrence were determined by retrospective review of medical records five years after initial surgery. Categories of recurrence were correlated to spectral intensities to explore potential metabolomic markers of recurrence in the cohort. Results Immunostainings with P504S and CK903 showed excellent staining quality and accessibility following 1H HRMAS MRS, suggesting these markers to be suitable for the presented quantitative approach to determine metabolomics profiles of PCa. In contrast, the quality of p63 IHC was impaired after previously performed spectroscopy. IHC using the immunomarkers P504S and CK903 on adjacent slides was found to present a feasible quantitative diagnostic method to distinguish between benign and cancerous conditions in prostate tissue. The cancer epithelium percentage as determined by QIAP showed a significant correlation to the results of quantitative IHC analysis performed by a pathologist (p < 0.001), as well as to a quantitative conventional histopathology review (p = 0.001). The same was true for the benign epithelium percentage (p < 0.001 and p = 0.0183), validating the presented approach. Two metabolomic regions showed a significant correlation between relative spectral intensities and the cancer epithelium percentage as determined by QIAP: 3.22 ppm (p = 0.015) and 2.68 ppm (p = 0.0144). The metabolites corresponding to these regions, phosphocholine and citrate, could be identified as metabolomic markers of PCa in the present cohort. 45 patients were followed for more than 12 months. Of these, 97.8% were still alive five years after initial surgery. 11 patients (24.4%) experienced a recurrence during the follow- up time. The categories of recurrence showed a correlation to the spectral intensities of two regions, 2.33 – 2.3 ppm (p = 0.0403) and 1.28 ppm (p = 0.0144), corresponding to the metabolites phosphocreatine and lipids. Conclusion This study introduces a method that allows an observer-independent, quantitative analysis of IHC to help establish metabolomic profiles and identify metabolomic markers of PCa from spectral intensities obtained with 1H HRMAS NMR Spectroscopy. The immunomarkers P504S and CK903 have been found suitable IHC analysis following 1H HRMAS MRS. A prospective in vivo application of PCa metabolite profiles and metabolomic markers determined by the presented method could serve as highly sensitive, non- invasive diagnostic tool. This observer- independent, computer- automated, quantitative analysis could help to distinguish highly aggressive tumors from low-malignant conditions, avoid overtreatment and reduce risks and complications for cancer patients in the future. Further studies are needed to verify the identified PCa metabolomic markers and to establish clinical applicability.:Table of Contents Glossary 1 Introduction 1. 1 Prostate Cancer 1. 2 Detection of Prostate Cancer – State of the Art 1. 2. 1 Prostate- Specific Antigen Test and Digital Rectal Examination 1.2.2 Radiographic Methods in PCa Detection 1.2.3 Transrectal Core Biopsies and Histopathological Analysis 1.2.4 Histopathological Grading of Prostate Cancer: GLEASON Score 1.3 Challenges and Need for New Approaches in PCa Diagnostic Management 2 Scientific Background I: Nuclear Magnetic Resonance,1H HRMAS NMR Spectroscopy and Metabolomic Profiles 2.1 Nuclear Magnetic Resonance 2.1.1 Spin Precession 2.1.2 Magnetic Resonance 2.1.3 Chemical Shift and J- coupling 2.2 Nuclear Magnetic Resonance 2.2.1 Magic Angle Spinning and 1H HRMAS NMR Spectroscopy 2.2.2 MAS Spinning Rates and Spinning Side Bands 2. 3 Metabolomics, Metabolite Profiles and Clinical Utility 3 Scientific Background II: Immunohistochemistry of Prostate Cancer 4 Aims of the Study 5 Material and Methods 5.1 Prostate Tissue Samples and Patient Demographics 5.2 1H HRMAS NMR Spectroscopy 5.2.1 Sample Preparation 5.2.2 Spectroscopy Scan 5.2.3 Data Processing 5.3 Immunohistochemistry 5.3.1 Immunohistochemistry Material and Equipment 5.3.2. Immunohistochemistry Protocol 5. 3. 3 Prostate Immunomarker Stability after 1H HRMAS NMR Spectroscopy 5.3.4 Qualitative IHC Analysis 5. 3.5 Quantitative IHC Analysis 5.3.5.1 Quantitative IHC Slide Review 5.3.5.2 Computer-Automated Quantitative IHC Analysis 5.3 Quantitative Histopathology 5. 4 Identification of Prostate Cancer Metabolomic Markers 5. 5 Patient Outcomes and Recurrence Categories 5.6 Statistical Analysis 6 Results 6. 1 Patient demographics 6. 2 Spectroscopy Results 6. 3 Immunohistochemistry 6. 3. 1 Evaluation of Prostate Immunomarker Stability after 1H HRMAS MRS 6. 3. 2 Qualitative Immunohistochemistry 6. 4 Quantitative Immunohistochemistry 6. 4. 1 Quantitative IHC Slide Review 6. 4. 2 Computer-Automated Quantitative IHC Evaluation using QIAP 6. 5 Quantitative Histopathology 6. 6 Identification of Prostate Cancer Metabolomic Markers using QIAP 6. 7 Patient Outcomes and Recurrence 7 Discussion 8 Summary / Abstract 9 Zusammenfassung 10 References 11 Erklärung über die eigenständige Abfassung der Arbeit 12 Danksagung 13 Lebenslauf und Publikationsverzeichnis Appendix A.1 Immunostaining protocols A.2 Spectral Intensities Measured by 1H HRMAS MRS in 51 Samples A.3 Graphs for Correlations of Spectral Intensities and CaE% determined by QIAP in 34 Additional Regions of Interest
Einführung Prostatakrebs ist eine häufigsten Krebserkrankungen in den USA und die zweithäufigste malignom- assoziierte Todesursache männlicher Patienten weltweit. Seit der Einführung des Prostata- spezifischen Antigen (PSA)- Screeningtests wird diese Krebsart in früheren Stadien diagnostiziert und therapiert, wodurch die Mortalitätsrate in den letzten Jahren deutlich reduziert werden konnte. Da moderne diagnostische Methoden bislang jedoch nicht ausreichend in der Lage sind, suffizient zwischen hochmalignen und weniger aggressiven Varianten dieses bösartigen Krebsleidens zu unterscheiden, werden häufig auch Patienten aggressiv therapiert, deren niedriggradiges Prostatakarzinom keine klinische Relevanz gehabt hätte. Es besteht daher ein großes wissenschaftliches Interesse an der Entwicklung neuer diagnostischer Methoden zur akkuraten Bestimmung von biologischem Status, Malignität, Aggressivität und Ausmaß einer Prostatakrebserkrankung. \\\\\\\"1H High Resolution Magic Angle Spinning Nuclear Magnetic Resonance Spectroscopy\\\\\\\" (1H HRMAS MRS) ist eine vielversprechende diagnostische Methode, welche es ermöglicht, metabolomische Profile von Prostatakrebs zu erstellen, ohne die Gewebsstruktur der analysierten Proben zu zerstören. Durch anschließende histopathologische Begutachtung lassen sich die erstellten Metabolitprofile validieren und evaluieren. Im Gegensatz zu konventionellen histopathologischen Methoden können durch immunhistochemische Verfahren dabei objektivere, akkuratere und quantifizierbare histopathologische Erkenntnisse gewonnen werden. Die vorliegende Studie präsentiert einen neuentwickelten diagnostischen Ansatz zur quantitativen Bestimmung von metabolomischen Markern von Prostatakrebs, basierend auf der Durchführung von 1H HRMAS NMR Spektroskopie und quantitativer Immunhistochemie. Material und Methoden Einundfünfzig Gewebsproben von Prostatakrebspatienten wurden mittels 1H HRMAS MRS an einem 14.1 T BRUKER NMR Spektrometer unter Einsatz einer CPMG-Pulssequenz untersucht. Spektrale Intensitäten in 36 Metabolitregionen wurden gemessen. Anschließend wurden die analysierten Gewebeproben mit drei Immunfärbemarkern für sowohl malignes (P504S, Alpha-methylacyl-CoA-racemase) als auch benignes (CK903, High-molecular weight cytokeratin, und p63) Prostatagewebe angefärbt und quantitativ mit Hilfe eines Bildanalyseprogramms (QIAP) ausgewertet. Die Anwendbarkeit und Auswertbarkeit der genannten Immunomarker nach Spektroskopie wurde evaluiert und mit der Färbungsqualität von nicht- gescannten Schnitten verglichen. Die Resultate der automatischen Auswertung durch QIAP konnten durch einen erfahrenen Pathologen in einer quantitativen Analyse der Immunfärbungen sowie konventioneller histologischer Färbungen derselben Gewebsproben validiert werden. Die spektralen Intensitäten aus den Messungen mit 1H HRMAS MRS wurden mit den korrespondierenden Ergebnissen der quantitativen Auswertung der Immunfärbungen korreliert, um metabolomische Marker von Prostatakrebs zu identifizieren. Der klinische Verlauf und die Rezidivrate der Patienten wurden 5 Jahre nach der initialen Prostatektomie retrospektiv bestimmt. Rezidivkategorien wurden erstellt und mit den bestimmten spektralen Intensitäten korreliert, um metabolomische Marker für das Auftreten von Prostatakrebsrezidiven zu identifizieren. Ergebnisse Die Immunfärbungen mit P504S und CK903 zeigten exzellente Qualität und Auswertbarkeit nach vorheriger 1H HRMAS MRS. Beide Marker eigneten sich zur Durchführung von quantitativer Immunhistochemie an spektroskopierten Gewebeproben. Im Gegensatz dazu war die Qualität der Immunfärbungen mit p63 nach Spektroskopie vermindert. Quantitative Immunfärbungen unter Einsatz der Immunmarker P504S und CK903 stellten eine praktikable diagnostische Methode dar, um zwischen malignen und benignem Prostatagewebe zu unterscheiden. Der Anteil von bösartig verändertem Prostatagewebe, bestimmt durch QIAP, korrelierte signifikant mit den Ergebnissen der quantitativen Analyse der Immunfärbungen durch den Pathologen (p < 0.001), sowie mit der quantitativen Auswertung der konventionellen histopathologischen Färbung (p = 0.001). Ebenso ließ sich die Bestimmung des Anteils von benignem Gewebe mit QIAP zu den Ergebnissen der pathologischen Analyse korrelieren (p < 0.001 und p = 0.0183). Für zwei metabolomische Regionen konnte ein signifikante Korrelation zwischen relativen spektralen Intensitäten, bestimmt mit 1H HRMAS NMR Spektroskopie, und dem Anteil von malignem Epithelium in derselben Gewebeprobe, ermittelt durch QIAP, festgestellt werden: 3.22 ppm (p = 0.015) und 2.68 ppm (p = 0.0144). Die zu diesen Regionen korrespondierenden Metaboliten, Phosphocholin und Zitrat, konnten als potentielle metabolomische Marker für Prostatakrebs identifiziert werden. Die retrospektiven Analyse der klinischen Daten der Patienten fünf Jahre nach Prostatektomie ergab eine Überlebensrate von 97.8%. Elf dieser Patienten (24.4%) erlitten ein Rezidiv ihrer Erkrankung. Die bestimmten Rezidivkategorien korrelierten signifikant mit zwei metabolomischen Regionen (2.33 – 2.3 ppm, p = 0.0403 und 1.28 ppm, p = 0.0144), welche zu den Metaboliten Phosphokreatin und Lipiden korrespondierten. Schlussfolgerung Die vorliegende Studie präsentiert einen diagnostischen Ansatz zur objektiven und quantitativen Bestimmung metabolomischer Marker von Prostatakrebs unter Verwendung von 1H HRMAS MRS und Immunhistochemie. P504S und CK903 eignen sich als Immunmarker für quantitative Immunfärbungen nach vorheriger Durchführung von 1H HRMAS MRS. Die Metaboliten Phosphocholin und Zitrat konnten in der vorliegenden Patientenkohorte als potentielle metabolomische Marker für Prostatakrebs identifiziert werden. Eine mögliche in vivo Anwendung der gefundenen metabolomischen Marker könnte als hochsensitives, objektives und nicht- invasives diagnostisches Werkzeug der Prostatakrebsdiagnostik dienen. Der vorliegende untersucherunabhängige, automatisierte und quantitative diagnostischer Ansatz hat das Potential, zwischen hochmalignen und weniger aggressiven Krebsfällen zu unterscheiden und somit unnötige Risiken und Komplikationen für Prostatakrebspatienten zu reduzieren. Weitere Untersuchungen sind notwendig, um die identifizierten metabolomischen Marker zu verifizieren und eine klinische Anwendung zu etablieren.:Table of Contents Glossary 1 Introduction 1. 1 Prostate Cancer 1. 2 Detection of Prostate Cancer – State of the Art 1. 2. 1 Prostate- Specific Antigen Test and Digital Rectal Examination 1.2.2 Radiographic Methods in PCa Detection 1.2.3 Transrectal Core Biopsies and Histopathological Analysis 1.2.4 Histopathological Grading of Prostate Cancer: GLEASON Score 1.3 Challenges and Need for New Approaches in PCa Diagnostic Management 2 Scientific Background I: Nuclear Magnetic Resonance,1H HRMAS NMR Spectroscopy and Metabolomic Profiles 2.1 Nuclear Magnetic Resonance 2.1.1 Spin Precession 2.1.2 Magnetic Resonance 2.1.3 Chemical Shift and J- coupling 2.2 Nuclear Magnetic Resonance 2.2.1 Magic Angle Spinning and 1H HRMAS NMR Spectroscopy 2.2.2 MAS Spinning Rates and Spinning Side Bands 2. 3 Metabolomics, Metabolite Profiles and Clinical Utility 3 Scientific Background II: Immunohistochemistry of Prostate Cancer 4 Aims of the Study 5 Material and Methods 5.1 Prostate Tissue Samples and Patient Demographics 5.2 1H HRMAS NMR Spectroscopy 5.2.1 Sample Preparation 5.2.2 Spectroscopy Scan 5.2.3 Data Processing 5.3 Immunohistochemistry 5.3.1 Immunohistochemistry Material and Equipment 5.3.2. Immunohistochemistry Protocol 5. 3. 3 Prostate Immunomarker Stability after 1H HRMAS NMR Spectroscopy 5.3.4 Qualitative IHC Analysis 5. 3.5 Quantitative IHC Analysis 5.3.5.1 Quantitative IHC Slide Review 5.3.5.2 Computer-Automated Quantitative IHC Analysis 5.3 Quantitative Histopathology 5. 4 Identification of Prostate Cancer Metabolomic Markers 5. 5 Patient Outcomes and Recurrence Categories 5.6 Statistical Analysis 6 Results 6. 1 Patient demographics 6. 2 Spectroscopy Results 6. 3 Immunohistochemistry 6. 3. 1 Evaluation of Prostate Immunomarker Stability after 1H HRMAS MRS 6. 3. 2 Qualitative Immunohistochemistry 6. 4 Quantitative Immunohistochemistry 6. 4. 1 Quantitative IHC Slide Review 6. 4. 2 Computer-Automated Quantitative IHC Evaluation using QIAP 6. 5 Quantitative Histopathology 6. 6 Identification of Prostate Cancer Metabolomic Markers using QIAP 6. 7 Patient Outcomes and Recurrence 7 Discussion 8 Summary / Abstract 9 Zusammenfassung 10 References 11 Erklärung über die eigenständige Abfassung der Arbeit 12 Danksagung 13 Lebenslauf und Publikationsverzeichnis Appendix A.1 Immunostaining protocols A.2 Spectral Intensities Measured by 1H HRMAS MRS in 51 Samples A.3 Graphs for Correlations of Spectral Intensities and CaE% determined by QIAP in 34 Additional Regions of Interest

Greyia radlkoferi is a South African indigenous tree, which has recently been discovered to be a source of extracts that have a potential in the development of cosmeceutical herbal products having the ability to treat hyperpigmentation disorders. For product development however, G. radlkoferi would need to be available in a commercial scale. Greyia radlkoferi grows naturally in the wild and is often available for cultivation as an ornamental plant. In order to establish this plant into cultivation, suitable propagation techniques must be established for rapid multiplication of trees and thus a sustainable leaf production. For consistent and improved leaf supply to the market, agronomic practices that will enhance leaf production were investigated in the current study. Furthermore, in order to meet market demand in terms of good quality extracts with guaranteed therapeutic efficiency, pre-harvest and post-harvest factors that affect the chemical composition of the extracts were investigated. Recently developed biotechnology techniques such as metabolomics using 1H-NMR and multivariate data analysis offered a platform to study the chemical variation of extracts. Therefore, the current study was aimed at understanding the requirements for propagation and optimum leaf production as well as conditions that favour optimum production of secondary metabolite of G. radlkoferi plant material (at pre and post-harvest) and thus assess its commercial viability. To understand the effects of temperature on seed germination of G. radlkoferi, seeds were exposed to five temperatures (10°C, 15°C, 20°C, 25°C and 30°C) in the incubators in the laboratory. Germination of G. radlkoferi by seeds was discovered to be temperature dependent. The optimum germination temperature of 81% was obtained at 25°C. In the case of vegetative propagation by stem cuttings, the effect of cutting position (basal or apical), exogenous rooting hormone (Seradix1, Seradix 2, 0.1% IBA, 0.3% IBA and 0.8% IBA) and cutting position were investigated in the glasshouse. The cutting position had a significant effect on rooting of G. radlkoferi cuttings with basal cuttings exhibiting 35% rooting as compared to 6% rooting attained for the apical cuttings. A clear trend in rooting response to application of rooting hormones was observed, with 0.1% Indole butyric acid (IBA) showing the highest rooting percentage of 63%. Considering the outcomes of the propagation studies as well as the limited material for vegetative propagation, seed propagation appears to be the most suitable technique for large-scale multiplication of G. radlkoferi. The effect of different pruning techniques as well as harvesting frequencies on fresh and dry weights of G. radlkoferi leaves were evaluated. Factors considered were four pruning treatments (‘pruned but not tipped’, ‘tipped but not pruned’, ‘not pruned nor tipped’ as well as ‘pruned and tipped’) and three harvesting periods (monthly, bimonthly and once–off). Bimonthly harvests highly increased leaf production compared to trees that were harvested monthly and once-off with higher leaf fresh weight yield of 238 g per tree or 2.38 tons/per hectare as well as dry weight yield of 83 g per tree or 0.830 tons/hectare. This outcomes of this study further suggested that a suitable pruning practice for G. radlkoferi would be to either ‘prune only’ or ‘cut back the main stem’ rather than a combination of the two treatments. The influence of seasons (summer, autumn, winter and spring) on the anti-tyrosinase activity and metabolomics profile of G. radlkoferi leaf extracts were investigated. Seasons significantly influenced the chemical composition and the efficacy of the plant extracts. Tyrosinase enzyme inhibition was investigated against monophenolase (tyrosine) with kojic acid as positive control. The highest tyrosinase inhibition concentration with IC50 (50% tyrosinase inhibition concentration) value of 30.3±1.8 μg/ml were obtained in winter harvested leaves compared to the other seasons. The lowest IC50 values were obtained in spring. Metabolomics analysis using orthogonal partial least square discriminant analysis (OPLS-DA) provided a clear class separation according to the harvest season. Extracts from winter harvested leaves contained sucrose, acetamide, alanine and a compound of the catechin group (gallocatechin-(4 alpha->8)-epigallocatechin) as revealed by 1H-NMR metabolomics with assistance of LC-MS. Since compounds of the catechin group are well-known tyrosinase inhibitors, the high tyrosinase activity exhibited in extracts of winter harvested G. radlkoferi leaves could be ascribed to the presence of gallocatechin-(4 alpha->8)-epigallocatechin. Based on the outcomes of the seasonal study, we suggest that in order to obtain extracts with high bioactivity, the best suitable time for harvesting leaf samples is in late autumn-early winter. Processing leaf material using three different drying methods (sun, oven and air drying) significantly influenced chemical composition and the efficacy of the plant extracts. Extracts prepared from air-dried leaf material showed the highest tyrosinase inhibition with IC50 value of 17.80 μg/ml compared to extracts of the other drying methods. Extracts of leaves processed with air drying preserved most metabolites during processing while extracts of sun-dried and oven-dried leaves clearly depleted some metabolites especially amino acids and some aromatic compounds. 1H-NMR metabolomics approach with the assistance of LC-MS data successfully determined a positive association of alanine, acetamide, sucrose and gallocatechin-(4 alpha->8)-epigallocatechin as the chemical constituents contributing to the variation in the air-dried leaves compared to the oven-dried leaves. A positive association of valine, alanine, leucine, isoleucine, gallocatechin-(4 alpha->8)-epigallocatechin and glucose contributed to the variation in air-dried group, compared to the sun-dried group. The highest tyrosinase inhibitory activity exhibited in air-dried samples compared to the other drying methods was associated with the presence of gallocatechin-(4 alpha->8)-epigallocatechin. Because air drying preserved most leaf metabolites compared to sun and oven drying, it was regarded as the most suitable method for processing G. radlkoferi leaf material. This study is the first scientific account that provides guidelines and recommendations to (1) establish G. radlkoferi as a cultivated plant for commercialization, (2) optimize leaf production for sustainable supply to the commercial markets and (3) optimize medicinal content of G. radlkoferi related to harvesting time and post-harvest processing (drying), for enhanced quality of extracts and its products
Agriculture, Animal Health and Human Ecology
Ph. D. (Agriculture)

碩士
國立臺灣大學
資訊網路與多媒體研究所
103
This dissertation presents two developed algorithms for solving computational problems of detecting small molecules in the field of metabolomics analysis. In the first part of this dissertation, we present the tool – LAKE, which is a tool for detected peak alignment to align retention time for chromatographic methods coupled to spectrophotometers such as high performance liquid chromatography for metabolomics works. The existed tools for retention time correction still can’t properly aligning retention times of detected peaks from multiple batches and some detected peaks are left misalignment. LAKE resolves peak shifts from high data similarity to low data similarity. In each turn, detected peaks would be clustered in mass-over-charged (m/z) dimension and then retention time (RT) dimension. For each m/z-RT cluster, bandwidth used in RT density estimation with kernel density estimation (KDE) is estimated with bandwidth selector. At the end of each turn of retention time shift resolution, the m/z and RT of detected peaks would be updated with average m/z and average RT of the m/z-RT group before next turn of detected peak alignment. LAKE can be applied to aligning retention time from mixed exogenic compounds samples, multiple exogenic compounds added in biofluid samples and complicate endogenous compounds contained metabolomics samples in multiple batches. In the second part of this dissertation, we present the tool – PHASION, which is a tool for automatic phase correction on multiple 1D proton nuclear magnetic resonance (1H-NMR) spectra for metabolomics works. The phase error is an unavoidable error happened when FID signal is recorded, after Fourier transformed into spectrum mixed with phase error. The phase correction is to find zeroth-order and first-order phase error to make misphased spectrum into phase-corrected spectrum before any further data processing. Current 1D 1H-NMR phase correction methods usually require manual parameter and filter tuning by experienced users to obtain desirable results from complex metabolomics spectra – thus becoming prone to correction variation and biased quantification. We present a novel alternative method, PHASION, for automatically estimating the phase angles of 1D 1H-NMR metabolomics data. PHASION finds optimal phase angles by calculating proposed objective score for relative stable segments of spectrum and calculates the score for baseline of spectrum phased with phase angles (PH0, PH1) and approach to the optimal phase angles for the spectrum with Nelder-Mead Simplex Optimizer.